The objective of the proposed research is to develop new theory and methods for high-resolution mapping of the mutations underlying complex genetic diseases of humans. The methods will use information from population level linkage disequilibrium (LD) between one or more genetic markers and a disease phenotype. The basis for LD mapping is that the marker haplotype found on the chromosome on which a disease mutation first arose will tend to have a higher frequency on chromosomes carrying that mutation. This non-random association between genetic markers and a disease phenotype decreases over time, due to recombination, at a rate that depends on the physical distance of the marker from the mutation. This allows the distances between linked genetic markers and a disease mutation to be estimated using a mathematical model of the process of genetic recombination and of the disease allele dynamics in the population. LD mapping methods are most effective when applied to populations that have experienced recent founding events and/or high levels of population growth. New statistical methods for LD mapping will be developed that make use of multiple genetic markers and account for potential complications such as recurrent mutation (the same disease phenotype results from different mutations in a single disease gene) and locus heterogeneity (the same disease phenotype results from mutations in several different disease genes). The statistical performance of the methods will be studied by using computer simulation to generate artificial populations of disease alleles and by analyzing existing genetic marker data for which the location of the disease mutation is now known. The methods will be suitable for use with several different types of genetic markers that are currently widely used including restriction fragment length polymorphisms (RFLPs), microsatellites, and single nucleotide polymorphisms (SNPs).